]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/drtv.py
Prepare to upload
[youtubedl] / youtube_dl / extractor / drtv.py
index d5df18d7c971c18f01c51128c75fbe4ee09ea070..f757745ba7cf2fa500f03102d7ca2b0c932c27bc 100644 (file)
@@ -1,34 +1,80 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
-from .subtitles import SubtitlesInfoExtractor
-from .common import ExtractorError
-from ..utils import parse_iso8601
-
-
-class DRTVIE(SubtitlesInfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dr\.dk/tv/se/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
-
-    _TEST = {
-        'url': 'http://www.dr.dk/tv/se/partiets-mand/partiets-mand-7-8',
-        'md5': '4a7e1dd65cdb2643500a3f753c942f25',
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    float_or_none,
+    mimetype2ext,
+    parse_iso8601,
+    remove_end,
+    update_url_query,
+)
+
+
+class DRTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv/se|nyheder|radio/ondemand)/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
+    _GEO_BYPASS = False
+    _GEO_COUNTRIES = ['DK']
+    IE_NAME = 'drtv'
+    _TESTS = [{
+        'url': 'https://www.dr.dk/tv/se/boern/ultra/klassen-ultra/klassen-darlig-taber-10',
+        'md5': '7ae17b4e18eb5d29212f424a7511c184',
         'info_dict': {
-            'id': 'partiets-mand-7-8',
+            'id': 'klassen-darlig-taber-10',
             'ext': 'mp4',
-            'title': 'Partiets mand (7:8)',
-            'description': 'md5:a684b90a8f9336cd4aab94b7647d7862',
-            'timestamp': 1403047940,
-            'upload_date': '20140617',
-            'duration': 1299.040,
+            'title': 'Klassen - Dårlig taber (10)',
+            'description': 'md5:815fe1b7fa656ed80580f31e8b3c79aa',
+            'timestamp': 1471991907,
+            'upload_date': '20160823',
+            'duration': 606.84,
         },
-    }
+    }, {
+        # embed
+        'url': 'https://www.dr.dk/nyheder/indland/live-christianias-rydning-af-pusher-street-er-i-gang',
+        'info_dict': {
+            'id': 'christiania-pusher-street-ryddes-drdkrjpo',
+            'ext': 'mp4',
+            'title': 'LIVE Christianias rydning af Pusher Street er i gang',
+            'description': 'md5:2a71898b15057e9b97334f61d04e6eb5',
+            'timestamp': 1472800279,
+            'upload_date': '20160902',
+            'duration': 131.4,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # with SignLanguage formats
+        'url': 'https://www.dr.dk/tv/se/historien-om-danmark/-/historien-om-danmark-stenalder',
+        'info_dict': {
+            'id': 'historien-om-danmark-stenalder',
+            'ext': 'mp4',
+            'title': 'Historien om Danmark: Stenalder (1)',
+            'description': 'md5:8c66dcbc1669bbc6f873879880f37f2a',
+            'timestamp': 1490401996,
+            'upload_date': '20170325',
+            'duration': 3502.04,
+            'formats': 'mincount:20',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
 
+        if '>Programmet er ikke længere tilgængeligt' in webpage:
+            raise ExtractorError(
+                'Video %s is not available' % video_id, expected=True)
+
         video_id = self._search_regex(
-            r'data-(?:material-identifier|episode-slug)="([^"]+)"',
+            (r'data-(?:material-identifier|episode-slug)="([^"]+)"',
+                r'data-resource="[^>"]+mu/programcard/expanded/([^"]+)"'),
             webpage, 'video id')
 
         programcard = self._download_json(
@@ -36,9 +82,12 @@ class DRTVIE(SubtitlesInfoExtractor):
             video_id, 'Downloading video JSON')
         data = programcard['Data'][0]
 
-        title = data['Title']
-        description = data['Description']
-        timestamp = parse_iso8601(data['CreatedTime'])
+        title = remove_end(self._og_search_title(
+            webpage, default=None), ' | TV | DR') or data['Title']
+        description = self._og_search_description(
+            webpage, default=None) or data.get('Description')
+
+        timestamp = parse_iso8601(data.get('CreatedTime'))
 
         thumbnail = None
         duration = None
@@ -49,45 +98,69 @@ class DRTVIE(SubtitlesInfoExtractor):
         subtitles = {}
 
         for asset in data['Assets']:
-            if asset['Kind'] == 'Image':
-                thumbnail = asset['Uri']
-            elif asset['Kind'] == 'VideoResource':
-                duration = asset['DurationInMilliseconds'] / 1000.0
-                restricted_to_denmark = asset['RestrictedToDenmark']
-                spoken_subtitles = asset['Target'] == 'SpokenSubtitles'
-                for link in asset['Links']:
-                    target = link['Target']
-                    uri = link['Uri']
-                    format_id = target
-                    preference = -1 if target == 'HDS' else -2
-                    if spoken_subtitles:
-                        preference -= 2
-                        format_id += '-spoken-subtitles'
-                    formats.append({
-                        'url': uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43' if target == 'HDS' else uri,
-                        'format_id': format_id,
-                        'ext': link['FileFormat'],
-                        'preference': preference,
-                    })
+            kind = asset.get('Kind')
+            if kind == 'Image':
+                thumbnail = asset.get('Uri')
+            elif kind in ('VideoResource', 'AudioResource'):
+                duration = float_or_none(asset.get('DurationInMilliseconds'), 1000)
+                restricted_to_denmark = asset.get('RestrictedToDenmark')
+                asset_target = asset.get('Target')
+                for link in asset.get('Links', []):
+                    uri = link.get('Uri')
+                    if not uri:
+                        continue
+                    target = link.get('Target')
+                    format_id = target or ''
+                    preference = None
+                    if asset_target in ('SpokenSubtitles', 'SignLanguage'):
+                        preference = -1
+                        format_id += '-%s' % asset_target
+                    if target == 'HDS':
+                        f4m_formats = self._extract_f4m_formats(
+                            uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43',
+                            video_id, preference, f4m_id=format_id, fatal=False)
+                        if kind == 'AudioResource':
+                            for f in f4m_formats:
+                                f['vcodec'] = 'none'
+                        formats.extend(f4m_formats)
+                    elif target == 'HLS':
+                        formats.extend(self._extract_m3u8_formats(
+                            uri, video_id, 'mp4', entry_protocol='m3u8_native',
+                            preference=preference, m3u8_id=format_id,
+                            fatal=False))
+                    else:
+                        bitrate = link.get('Bitrate')
+                        if bitrate:
+                            format_id += '-%s' % bitrate
+                        formats.append({
+                            'url': uri,
+                            'format_id': format_id,
+                            'tbr': int_or_none(bitrate),
+                            'ext': link.get('FileFormat'),
+                            'vcodec': 'none' if kind == 'AudioResource' else None,
+                            'preference': preference,
+                        })
                 subtitles_list = asset.get('SubtitlesList')
                 if isinstance(subtitles_list, list):
                     LANGS = {
-                        'Danish': 'dk',
+                        'Danish': 'da',
                     }
                     for subs in subtitles_list:
-                        lang = subs['Language']
-                        subtitles[LANGS.get(lang, lang)] = subs['Uri']
+                        if not subs.get('Uri'):
+                            continue
+                        lang = subs.get('Language') or 'da'
+                        subtitles.setdefault(LANGS.get(lang, lang), []).append({
+                            'url': subs['Uri'],
+                            'ext': mimetype2ext(subs.get('MimeType')) or 'vtt'
+                        })
 
         if not formats and restricted_to_denmark:
-            raise ExtractorError(
-                'Unfortunately, DR is not allowed to show this program outside Denmark.', expected=True)
+            self.raise_geo_restricted(
+                'Unfortunately, DR is not allowed to show this program outside Denmark.',
+                countries=self._GEO_COUNTRIES)
 
         self._sort_formats(formats)
 
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, subtitles)
-            return
-
         return {
             'id': video_id,
             'title': title,
@@ -96,5 +169,61 @@ class DRTVIE(SubtitlesInfoExtractor):
             'timestamp': timestamp,
             'duration': duration,
             'formats': formats,
-            'subtitles': self.extract_subtitles(video_id, subtitles),
+            'subtitles': subtitles,
+        }
+
+
+class DRTVLiveIE(InfoExtractor):
+    IE_NAME = 'drtv:live'
+    _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv|TV)/live/(?P<id>[\da-z-]+)'
+    _GEO_COUNTRIES = ['DK']
+    _TEST = {
+        'url': 'https://www.dr.dk/tv/live/dr1',
+        'info_dict': {
+            'id': 'dr1',
+            'ext': 'mp4',
+            'title': 're:^DR1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        channel_id = self._match_id(url)
+        channel_data = self._download_json(
+            'https://www.dr.dk/mu-online/api/1.0/channel/' + channel_id,
+            channel_id)
+        title = self._live_title(channel_data['Title'])
+
+        formats = []
+        for streaming_server in channel_data.get('StreamingServers', []):
+            server = streaming_server.get('Server')
+            if not server:
+                continue
+            link_type = streaming_server.get('LinkType')
+            for quality in streaming_server.get('Qualities', []):
+                for stream in quality.get('Streams', []):
+                    stream_path = stream.get('Stream')
+                    if not stream_path:
+                        continue
+                    stream_url = update_url_query(
+                        '%s/%s' % (server, stream_path), {'b': ''})
+                    if link_type == 'HLS':
+                        formats.extend(self._extract_m3u8_formats(
+                            stream_url, channel_id, 'mp4',
+                            m3u8_id=link_type, fatal=False, live=True))
+                    elif link_type == 'HDS':
+                        formats.extend(self._extract_f4m_formats(update_url_query(
+                            '%s/%s' % (server, stream_path), {'hdcore': '3.7.0'}),
+                            channel_id, f4m_id=link_type, fatal=False))
+        self._sort_formats(formats)
+
+        return {
+            'id': channel_id,
+            'title': title,
+            'thumbnail': channel_data.get('PrimaryImageUri'),
+            'formats': formats,
+            'is_live': True,
         }